Professor Gina RipponThis is a guest post by Georgina Rippon, who is Professor of Cognitive Imaging and Pro-Vice Chancellor (International) at Aston University. Gina recently gave the ScienceGrrl Birmingham ‘Say NO to Neurotrash!’ lecture (video link). Find her on Twitter:  @springfield16.

We watched the latest discussion on ‘male and female brains’ play out in the media with dismay, and so ScienceGrrl’s Anna Zecharia turned to Gina to help sort the neuronews from the neurononsense. As she and her colleagues say in a recent paper: “studies tend simply to compare the biological sexes, as though the implicit aim were to identify fixed, universal female versus male signatures”. Isn’t it about time we had a more nuanced conversation?


Here we go again! A complex brain imaging study reports differences between men and women’s brains and the media frenzy begins, saying: “We were right all along, male and female brains are different!!”

Sometimes you can blame the media for mangling the findings of a complicated neuroimaging study where the basic premises are not well-spelled out or where the innumerable steps from scanner to “pretty brain picture” are obscured.  But sometimes the scientists themselves can carelessly, if not deliberately, contribute to the furore. I think this is the case here.

Regardless of media interpretation, there is a lot to critique in the study itself. First the good stuff: the technique itself and its application are sound and the group size is impressive. However, there may be problems with controls – and there would definitely have been better ways to analyse the data.

Diffusion Tensor Imaging (DTI) is a complex procedure (for a quick Wiki-guide, check here) that requires quite some time lying still in a scanner. It has been shown that moving around while in the scanner can have distorting effect on connectivity measures. What I’d really like to know is – did these researchers control for movement differences between their groups?

The authors point out that the observed differences depend on age – that is, they are not really evident until after age 13. However, I’m not sure they controlled well enough for developmental stage – they say their groups “correspond roughly to the developmental stages of childhood, adolescence and young adulthood….” This does not smack of rigorous scientific methodology to me! If these points were properly controlled for, they were not properly reported in the paper.

Which brings us to another stumbling block when it comes to controls: we don’t know much about the subjects. We know there are 949 of them, their age (8-22 years), their sex and (in the words of the authors) their ‘race’. Do we know anything about the life experiences of the people they measured? Have they been brought up in an all-male family? An all-female family? Mixed? What is their level of education and socioeconomic status?

Regarding the data analysis, they say they have collected behavioural data but make no attempt to incorporate these data in a full-scale analysis. The authors might have been better to carry out a regression analysis, a form of analysis which would allow them to look at the group as a whole and identify which characteristics best predicted the structural differences they found.

Then to the big one – if we could fully understand the structural differences, does this tell us anything about how these connections could be used? No. For me, the scientists’ biggest error is the massive leap from structure to function – this really is a cardinal sin in cognitive neuroscience circles. Before interpreting their findings in these terms, the scientists should have asked themselves the following questions:

*   Is bigger better?
*   Are more connections better than fewer?
*   Does this technique allow a direct measure of the relationship between structure and function?

What is being described here is a road map of connections, with the unchallenged suggestion that the bigger and better the connections the more efficient the traffic flow. Anyone ever been on the LA freeway system? Or, conversely, on a German Autobahn? This brain imaging technique would not be able to measure the kind of traffic flow they are claiming: this technique provides a static ‘snapshot’ of fixed pathways in the brain, giving no clue as to when or how they are used. Any relationship with behaviour can only be speculative at this stage.

What the study actually shows is that there are structural differences in brain connections between two groups, one male, one female. The differences are not absolute i.e. not all male brains showed one pattern, all females another. But let’s ignore that – it’s much more exciting to report that neuroscientists are proving the accuracy of stereotypes!

This time, the responsibility really does seem to lie with the scientists. Both in the paper:

“The observations suggest that male brains are structured to facilitate connectivity between perception and coordinated action, whereas female brains are designed to facilitate communication between analytical and intuitive processing modes.”

and in the press:

“I was surprised that it matched a lot of the stereotypes that we think we have in our heads”

the lead scientist Dr Ragini Verma has allowed the ‘story’ to take over the science – for ‘neuronews’ to become ‘neurononsense’. Challenging claims like this is where peer review and journalists should come in – before the genie is out of the bottle and harmful gender stereotypes are perpetuated once again.

The brain is much more plastic than the early neuroscientists ever dreamed, it is hugely permeable to society’s influences. Life experiences can (literally) be brain-changing – and any talk of hardwiring is to misunderstand neural development. Until social factors are controlled for, it is impossible to say any differences are solely due to gender. Our brains reflect the society we live in.


ScienceGrrl says:

The society we live in is one in which girls and women are under-represented in STEM. Recent results from OECD PISA 2012 show that girls across the world do not have confidence in their abilities in maths – this is perhaps unsurprising considering girls and young women are being made to feel patronised or stupid just because they are female. Girls have to fight traditional and limiting gender stereotypes in order to reconcile their identity with a love of STEM. Sensationalist reporting of studies like this cannot help matters.

Follow the reaction on Twitter: http://storify.com/deevybee/postpublication-peer-review-on-sex-differences-in?utm_campaign=website&utm_source=email&utm_medium=email

and continue the conversation using #neurotrash.